The discovery of reverse tricyclic pyridone JAK2 inhibitors. Part 2: Lead optimization

Article abstract of DOI:10.1016/j.bmcl.2014.02.011

This communication discusses the discovery of novel reverse tricyclic pyridones as inhibitors of Janus kinase 2 (JAK2). By using a kinase cross screening approach coupled with molecular modeling, a unique inhibitor-water interaction was discovered to impart excellent broad kinase selectivity. Improvements in intrinsic potency were achieved by utilizing a rapid library approach, while targeted structural changes to lower lipophilicity led to improved rat pharmacokinetics. This multi-pronged approach led to the identification of 31, which demonstrated encouraging rat pharmacokinetics, in vivo potency, and excellent off-target kinase selectivity.

Full text of DOI:10.1016/j.bmcl.2014.02.011

Bioorganic & Medicinal Chemistry Letters 24 (2014) 1466–1471
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The discovery of reverse tricyclic pyridone JAK2 inhibitors.
Part 2: Lead optimization
a
b
a
a
Tony Siu a, , Sathyajith E. Kumarasinghe , Michael D. Altman , Matthew Katcher , Alan Northrup ,
Catherine White a, Craig Rosenstein c, Anjili Mathur d, Lin Xu e, Grace Chan c, Eric Bachman d,
Melaney Bouthillette f, Christopher J. Dinsmore a, C. Gary Marshall g, Jonathan R. Young a
⇑
a Department of Medicinal Chemistry, Merck & Co., 33 Avenue Louis Pasteur, Boston, MA 02115, USA
b Department of Structural Chemistry, Merck & Co., 33 Avenue Louis Pasteur, Boston, MA 02115, USA
c Department of In Vitro Sciences, Merck & Co., 33 Avenue Louis Pasteur, Boston, MA 02115, USA
d Department of Pharmacology, Merck & Co., 33 Avenue Louis Pasteur, Boston, MA 02115, USA
e Department of Drug Metabolism, Merck & Co., 33 Avenue Louis Pasteur, Boston, MA 02115, USA
f Department of Basic Pharmaceutical Sciences, Merck & Co., 33 Avenue Louis Pasteur, Boston, MA 02115, USA
g Department of Oncology, Merck & Co., 33 Avenue Louis Pasteur, Boston, MA 02115, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
This communication discusses the discovery of novel reverse tricyclic pyridones as inhibitors of Janus
kinase 2 (JAK2). By using a kinase cross screening approach coupled with molecular modeling, a unique
inhibitor–water interaction was discovered to impart excellent broad kinase selectivity. Improvements in
intrinsic potency were achieved by utilizing a rapid library approach, while targeted structural changes to
lower lipophilicity led to improved rat pharmacokinetics. This multi-pronged approach led to the iden-
tification of 31, which demonstrated encouraging rat pharmacokinetics, in vivo potency, and excellent
off-target kinase selectivity.
Received 7 January 2014
Revised 31 January 2014
Accepted 4 February 2014
Available online 15 February 2014
Keywords:
JAK2
Kinases
Ó 2014 Elsevier Ltd. All rights reserved.
Water interaction
Kinase selectivity
Ligand binding affinity
Kinase partition index
Molecular modeling
The Janus kinases, (JAK1, JAK2, JAK3, and TYK2), are a family of
non-receptor tyrosine kinases that mediate cytokine signaling and
play a central role in regulating cell proliferation and immune
response.1 Most notably, JAK2 has attracted significant attention
due to the discovery of the JAK2 V617F gain of function mutation
and its causative link to polycythemia vera (PV), essential throm-
bocythemia (ET), and primary myelofibrosis (MF).2 These findings
prompted the initiation of numerous JAK2 medicinal chemistry
programs,3 leading to over 60 clinical investigations of JAK2 inhib-
itors and ultimately resulting in the approval of Ruxolitinib for
MF.4
In a previous communication, we reported the discovery and
optimization of a novel JAK2 tricyclic pyridone inhibitor 1 (Fig. 1)
from a screening hit to a lead compound suitable for in vivo proof
of concept.5 Our initial strategy focused on monitoring ligand bind-
ing efficiency (LBE) while relying on structural information in con-
junction with a rapid library approach to afford 1, which has
excellent LBE (0.49), intrinsic potency (enzyme JAK2 IC50 = 1 nM),6
cell potency (cell JAK2 IC50 = 40 nM),7 and demonstrated in vivo
PK/PD activity (IC50 = 2700 nM).8 Despite the excellent in vitro
and modest in vivo JAK2 potency, 1 and analogs within this series
of fused napthyridinones suffered from notable liabilities that re-
quired optimization before advancing further in development. In
order to minimize unforeseeable adverse effects and toxicity due
to broad inhibition of cellular processes, a stringent selectivity
window of 100 fold or greater against a broad spectrum of kinases
was required. Furthermore, despite the excellent in vitro potency
of 1, the in vivo properties of 1 required additional optimization
to achieve more efficient dosing. In this communication, we ex-
pand upon our initial SAR findings and detail the optimization of
kinase selectivity and in vivo properties leading to the identifica-
tion of 31.
Compound 1, although quite potent against JAK2, is also a po-
tent inhibitor of a variety of kinases as tested against a broad 98
kinase panel (89% of kinases >100 fold over JAK29 with a partition
⇑
Corresponding author. Tel.: +1 617 939 0216.
E-mail address: tony_siu@merck.com (T. Siu).
http://dx.doi.org/10.1016/j.bmcl.2014.02.011
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

T. Siu et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1466–1471
1467
HN
O
N
CF3
N
H
N
1
JAK2 IC50 = 1 nM
Cell JAK2 IC50 = 40 nM
Rat Cl = 68 mL/min/kg
PK/PD IC50 = 2700 nM
Figure 1. Lead identification hit for JAK2.
index10 of 0.56) (Fig. 3). Moreover, the kinase profiles of various
other related analogs did not offer improvements in kinase selec-
tivity, which is consistent with the modeled binding mode for this
class of inhibitors. As discussed in a previous communication,5 our
strategy for introducing chemical diversity was to utilize a varied
set of amines to fill the ribose pocket space (Fig. 2). However, fur-
ther diversification in this vector would presumably not signifi-
cantly alter the kinase selectivity to the desired levels due to the
high conservation of residues among kinases in this pocket. Fur-
thermore, we imagined that branching out of the enzyme into
the solvent front would also not improve selectivity since the
inhibitor would not make contact with the enzyme surface.
In order to cross-fertilize kinase programs and to leverage ser-
% Kinases >100X
over JAK2
Compounds
Partition Index10
1
89
94
98
0.56
0.70
0.92
14
31
endipity, Merck Research Laboratories implemented
a cross-
screening effort between JAK2 and other Merck kinase programs.
This cross-program collaboration led to the identification of 3,
where by the amino pyridine 2 was altered to its regio isomer
(Scheme 1).
Figure 3. Kinase selectivity comparison of compounds 1, 14, and 31 against 98
kinases using a radar plot, percent of kinases >100 fold over JAK2, and partition
index.10
While this subtle core modification did not affect intrinsic po-
tency, significant improvements in selectivity against a mini panel
of off-target kinases were observed. Based on this serendipitous
discovery, we were motivated to further expand upon this initial
finding. Because the chemistry lent itself to a divergent synthesis,
a library approach was utilized to rapidly generate diverse SAR.
The synthesis of the fused napthyridinones core of 3 was per-
formed in an analogous manner to 1 (Scheme 2).11 Starting with 2-
ethoxy nicotinic acid 4, chlorination with oxalyl chloride followed
by the addition of diisopropyl amine afforded amide 5. With the
diisopropyl amide in place, directed ortho lithiation with sec-BuLi
followed by trapping of the anion with trimethylborate furnished
boronic acid 6. Suzuki cross coupling of boronic acid 6 with commer-
cially available amino pyridine 7, catalyzed by tetrakis palladium (0)
triphenylphosphine, provided the coupled biaryl product 8. Base in-
duced ring closure with sodium bis(trimethylsilyl)amide (NaHMDS)
afforded 9 as a poorly soluble polyaromatic molecule. Compound 9
was then subsequently heated in neat phosphoryl chloride at
100 °C in a microwave reactor resulting in a mixture of double chlo-
HN
HN
H
N
O
O
N
N
N
H
F
F
2
3
JAK2 = 100 nM
JAK2 = 200 nM
Scheme 1. Reversal of amino pyridine leads to increased kinase selectivity.
rination 10b and monochlorination 10a intermediates. Late stage
intermediates 10a and 10b allowed for rapid analog generation,
and thus a set of amines were substituted thermally at 65 °C with
base assistance using sodium tert-butoxide. Final deprotection of
the ethoxy ether 11a with boron tribromide or acid hydrolysis of
the chloro 11b with 6 N HCl, yielded the pyridone analogs.
Table 1 summarizes the potency of a select set of amine and
anilines. The focus of this library was to improve on the intrinsic
JAK2 potency of 3 through enhanced contact with the enzyme
and to evaluate the overall kinase selectivity for this new class of
compounds. Initially, trifluroamine 13 was prepared based on the
favorable activity of 1. Although 13 was active in enzyme and cells,
the shift in potency with respect to 1 indicated that the SAR gener-
ated in our previous communication did not directly translate to
the reverse pyridine series. Gratifyingly, 2,6-disubstituted anilines
were quickly discovered to possess exquisite enzyme and cellular
potency, highlighted by a focused library (14–20). The increased
potency of the ortho aniline substituents of these compounds
underscored the importance of a conformationally twisted aniline
H
i
HN
n
g
e
Conserved
Ribose Pocket
O
N
CF3
N
H
N
Solvent Front
1
Figure 2. General binding of fused napthyridones to JAK2 kinase.

1468
T. Siu et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1466–1471
Table 1
pocket of the enzyme with the NH pointing towards the solvent
front.5 Compound 14 is proposed to bind to the hinge in a manner
similar to 1, with the aniline functional group making favorable
hydrophobic interactions in the ribose pocket and with the NH of
the aniline pointing towards the back pocket. Modeling suggests
that the reversed amino pyridine of 14 allows the NH to interact
with a bound water molecule (H2O-139 in PDB 2B7A) which in
turn forms hydrogen bonds with the backbone of residues
Gly-993 and Asp-994. The presence of this crystallographic water
molecule may be a consequence of the unique conformation of
Gly-993 where the backbone is flipped relative to the DFG-1 resi-
dues of most other kinases.13 This unique interaction possibly
helps to generate more specific affinity towards the JAK2 enzyme
in particular.enzyme in particular.14
SAR of fused napthyridinone analogs
HN
O
H
N
R
N
N
Compound
R
JAK2 IC50 (nM)
10
Cell JAK2 IC50 (nM)
170
R
13
In pharmacokinetic studies, 14 showed a moderate rat plasma
clearance (41 mL/min/kg) and volume (2.6 L/kg) and a short half-
life (0.8 h). Moreover, 14 was profiled in our in vivo PK/PD model,
where it displayed an IC50 of 2300 nM. Thus, despite advances in
the in vitro properties and the improved kinase selectivity of 14
as compared to 1, the modest rat Cl and in vivo potency high-
lighted the need for further optimization.
To address the pharmacokinetics and in vivo potency, we fo-
cused on improving the physicochemical properties by lowering
the lipophilicity, and hence the logD of 14. The reduction of intrin-
sic lipophilicity is central to our strategy to increase the stability of
our compounds towards oxidative metabolism.15 The computa-
tional docking model of 14 (Fig. 4) suggested that the aromatic
4-fluorine of the aniline might be amendable to structural changes
since it occupies an open pocket and does not come in contact with
specific interactions with the enzyme backbone. Modifications
were aimed at introducing polar functional groups to influence
physicochemical properties. Additionally, we noticed an opportu-
nity to possibly interact with the carbonyl of Asn-981, and hence
our designs for modifying logD in this area focused on heteroatoms
with hydrogen bond donating capabilities (Fig. 5).
To access this series of compounds with modifications at the
aniline 4-position, further synthetic transformations were needed
(Scheme 3).11 Starting with 1-(4-amino-3,5-dichlorophenyl)etha-
none 21, ketal formation under acidic conditions afforded the ketal
22. Masking the ketone was necessary in order to disrupt nitrogen
lone pair conjugation to the ketone and increase the nucleophilic-
ity of the aniline for addition to the tricyclic core. This protected
aniline was carried through to 23 using the same conditions as in
Scheme 2, followed by deprotection with boron tribromide. So-
dium borohydride reduction of the ketone provided secondary
alcohol 24.16 Alternatively, treatment of the ketone with excess
methyl magnesium bromide afforded the tertiary alcohol 25.
Additional methods to functionalize the para position of the aniline
ulitilized palladium cross coupling chemistry. 2,6-Dichloro-4-iodo-
aniline was introduced into tricyclic core 10a according to the
methods described in Scheme 2. Suzuki reaction of 27 with pyra-
zole-5-ylboronic acid followed by deprotection with boron tribro-
mide yielded 28. Additionally, 27 could be deprotected and cross
coupled with boron triflouride salt 2917 to afford 30 as a mixture
of ketone and ketals. This mixture was treated with sodium
borohydride to afford hydride to afford 3131.16
With ready access to substituted anilines in hand, we tested our
hypothesis of improving in vivo properties through manipulation
of the overall physicochemical property of the molecule (Table 2).
Analogs identified which maintained high JAK2 cell potency were
profiled in rat PK studies, followed by evaluation in our in vivo
PK/PD model.18 In order to compare PK properties among com-
pounds, we accounted for the differences in plasma protein bind-
ing by normalizing the rat total Cl to the rat unbound Cl, a key
parameter to assess improvements in PK. As an initial exploration
to decrease the logD, we tested sulfonamide 32. The introduction
CF3
F
R
14
15
16
17
18
19
20
0.9
7
15
90
14
24
140
49
90
F
Cl
R
F
F
F
Cl
R
Cl
R
Cl
R
Cl
R
Cl
R
F
0.8
2
F
Cl
Cl
Cl
N
3
OCF3
CF3
1
CF3
9
ring to enhance hydrophobic interactions in the ribose pocket and
glycine rich loop. In an optimized case, 2,4,6-tri-halogens were
preferred as exemplified by 14 (JAK2 IC50 = 0.9 nM and cell JAK2
IC50 = 15 nM).
Having optimized for potency, 14 was profiled in a broad kinase
panel consisting of 98 kinases and compared to 1 as shown in
Figure 3.9 Kinase selectivity is depicted using a radar plot where
the distance from the center is proportional to the fold selectivity
over JAK2 in log10 units. From the depiction of data in Figure 3, it
can be observed that 14 is more selective because its curve lies out-
side that of 1. Further metrics to highlight the improved selectivity
were the increased number of kinases greater than 100-fold over
JAK2 (94%) as well as an improved partition index10 (0.70) as com-
pared to 1.
In order to better understand the nature of the improved kinase
selectivity, inhibitors 1 and 14 were modeled into a JAK2 crystal
structure (Fig. 4).12 As previously discussed, the pyridone of 1 is
proposed to hydrogen bond to hinge residues Glu-930 and Leu-
932. Furthermore, the CF3 amine linkage may occupy the ribose

T. Siu et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1466–1471
1469
N
N
N
a,b
c
N
N
O
OH
O
O
B
O
O
HO
O
OH
5
6
4
NH2
I
N
O
N
N
7
N
OH
O
O
e
N
d
f
N
NH2
N
8
9
HN
N
N
N
H
H
N
N
Cl
O
R
Y
R
Y
g
h
N
N
N
N
N
10a
Y= OEt
11a
= OEt
12
10b Y= Cl
11b = Cl
Scheme 2. Synthesis of fused napthyridinones. Reagents and conditions: (a) oxalyl chloride, CH2C12; (b) diisopropyl amine, triethylamine, CH2C12; (c) sec-BuLi,
trimethylborate, À78 °C, THF; (d) 7, Pd(PPh3)4, 2.0 M Na2CO3, 100 °C, THF; (e) NaHMDS, 0 °C, THF; (f) POCl3, 100 °C, microwave; (g) sodium tert-butoxide, amines, 85 °C, THF;
(h) Y = OEt, BBr3, 85 °C, CHC13 or Y = Cl, 6 N HCl dioxane, 85 °C.
b
24
NH2
NH2
HN
Cl
Cl
Cl
H
N
Cl
Cl
Scheme 2
a
O
N
O
Cl
N
O
O
O
c
25
21
22
23
BF
3
F
O
O
O
HN
Cl
NH2
F
N
Cl
H
N
H
N
29
Cl
Cl
O
e
Scheme 2
EtO
31
N
CHF2
N
Cl
d
Cl
I
N
X
I
N
X = O, OH (OMe), (OMe)2
30
26
27
f,g
28
Scheme 3. Synthesis of fused napthyridinones with substituted anilines. Reagents and conditions: (a) ethylene glycol, PPTs, 100 °C, benzene; (b) NaBH4, MeOH; (c) (5 equiv)
1.0 M MeMgBr in THF/toluene, 0 °C, THF; (d) (i) Et3N, PdCl2(dppf)–CH2Cl2, 29, 90 °C, n-PrOH; (ii) BBr3, 85 °C, CHCl3; (e) NaBH4, MeOH; (f) Pd(PPh3)4, 2.0 M NaHCO3, 1H-
pyrazole-5-boronic acid, 80 °C, THF; (g) BBr3, 85 °C, CH°C, CH22ClCl2.

1470
T. Siu et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1466–1471
Table 2
SAR of 4-substituted anilines
HN
H
N
O
R
N
N
R
JAK2 enzyme/cell IC50 (nM)
0.9/15
HplcLogD
Rat PPB % bound
95.2
Rat CI (mL/min/kg)
42
Unbound rat CI (mL/min/kg)
875
PK/PD IC50 (nM)
2300
Cl
R
14
1.75
F
F
Cl
Cl
R
24
25
28
0.6/12
0.5/18
1/37
1.20
1.34
1.43
95.2
97.4
99.9
54
29
10
1125
400
1200
NA
Cl
OH
R
1115
Cl
OH
Cl
Cl
R
10,000
Cl
HN N
F
R
31
32
0.2/26
1.46
1.17
97.2
NA
16
50
571
NA
1300
NA
F
Cl
OH
Cl
R
0.15/113
NH2
O
S
Cl
O
of a sulfonamide moiety lowered the logD by half a log unit as
compared to 14 and increased the overall JAK2 affinity
(IC50 = 0.15 nM). This result further validated our hypothesis that
additional polarity could be introduced in this flexible region of
the kinase. However, the increased polarity dramatically affected
the cell shift by >700 fold. Further investigation with 2-pyrazole
28 resulted in lower logD while maintaining favorably potency;
however, this functional group offered no PK advantage (unbound
Cl = 10000 mL/min/kg) with respect to 14. Polar groups such as
alcohols were also investigated as attractive groups to lower lipo-
philicity. To that end, alcohol 24 was profiled and showed poor rat
PK (unbound Cl = 1125 mL/min/kg), but with a much improved
in vivo potency (PK/PD IC50 = 400 nM). Despite the poor rat un-
bound Cl, we were encouraged with the potency and favorable
logD associated with the alcohol functional group. We hypothe-
sized that the rapid metabolism of 24 could occur through the glu-
curonidation or oxidation of the alcohol functionality, and that
blocking or attenuating these events would be beneficial for
improving the rat pharmacokinetic properties. Consequently, both
25 and 31 were synthesized with the goal of attenuating the trans-
formation of the alcohol with steric and electronic features. Disap-
pointingly, tertiary alcohol 25 still suffered from high rat unbound
Cl (1115 mL/min/kg); however, difluoro methyl alcohol 31 had an
improved rat unbound Cl (571 mL/min/kg), validating our hypoth-
esis. Furthermore, 31 was tested in our PK/PD assay and displayed
a respectable in vivo potency of 1300 nM. Although modest po-
tency gains were made in vivo, when combined with the improved
rat unbound Cl, 31 possessed a more desirable in vivo profile than
14. Along with the improved in vivo properties, the addition of the
OH to Asn-981 interaction contributed to further increasing the ki-
nase selectivity of 31 with notable improvements compared to 1 in
the radar plot, 100-fold JAK2 selectivity over 98% of the kinase pa-
nel, and a partition index10 of 0.92.
In this communication, we detailed the strategy and approach
used to optimize our initial JAK2 leads into inhibitors with favor-
able drug like properties. By using a kinase cross-screening ap-
proach to leverage our internal network capabilities, we
discovered the reversed pyridine 3 as a critical SAR to improve ki-
nase selectivity. While a rapid library approach aided in identifying
analogs which improved intrinsic potency by 200-fold, molecular

T. Siu et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1466–1471
1471
Figure 4. Inhibitors 1 and 14 modeled in a JAK2 crystal structure showing the unique interaction of 14 with water-139.
K.; Dinsmore, C. J.; Altman, M. D.; Chan, G.; Rosenstein, C.; Sharma, S.; Su, H.-P.;
Szewczak, A. A.; Xu, L.; Yin, H.; Zugay-Murphy, J.; Marshall, C. G.; Young, J. R. J.
Med. Chem. 2011, 54, 7334; (e) Harikrishnan, L. S.; Kamau, M. G.; Wan, H.;
Inghrim, J. A.; Zimmermann, K.; Sang, X.; Mastalerz, H. A.; Johnson, W. L.;
Zhang, G.; Lombardo, L. J.; Poss, M. A.; Trainor, G. L.; Tokarski, J. S.; Lorenzi, M.
V.; You, D.; Gottardis, M. M.; Bladwin, K. F.; Lippy, J.; Nirschi, D. S.; Qui, R.;
Miller, A. V.; Khan, J.; Sack, J. S.; Purandare, A. V. Bioorg. Med. Chem. Lett. 2011,
21, 1425; (f) Labadie, S.; Barrett, K.; Blair, W. S.; Chang, C.; Deskmukh, G.;
Eigenbrot, C.; Gibbons, P.; Johnson, A.; Kenny, J. R.; Kohli, P. B.; Liimatta, M.;
Lupardus, P. J.; Shia, S.; Steffek, M.; Ubhayakar, S.; van Abbema, A.; Zak, M.
Bioorg. Med. Chem. Lett. 2013, 21, 5923.
4. Harry, B. L.; Eckhardt, S. G.; Jimeno, A. Expert Opin. Ther. Targets 2012, 21, 637.
5. Siu, T.; Kozina, E. S.; Jung, J.; Rosenstein, C.; Mathur, A.; Altman, M. D.; Chan, G.;
Xu, L.; Bachman, E.; Mo, J.; Bouthillette, M.; Rush, T.; Dinsmore, C. J.; Marshall,
C. G.; Young, J. R. Bioorg. Med. Chem. Lett. 2010, 20, 7421.
6. Detection of peptide phosphorylation was performed by homogenous time
resolved fluorescence (HTRF) using
a europium chelate. IC50 values are
reported as the averages of at least two independent determinations;
standard deviations are within +/À 25–50% of IC50 values. For further details
see: WO2009035575.
7. Detection of STAT phosphorylation was performed using AlphaScreen™
SureFire™ p-STAT5 assay (Perkin Elmer and TGR Biosciences) using both
biotinylated anti-phospho-STAT5 antibody, which is captured by Streptavidin-
coated Donor beads, and anti-total STAT5 antibody, which is captured by
Protein A conjugated Acceptor beads. IC50 values are reported as the averages
of at least two independent determinations; standard deviations are within +/À
25–50% of IC50 values. For further details see: WO2008156726.
8. Mathur, A.; Mo, J.; Kraus, M.; O’Hare, E.; Sinclair, P.; Young, J.; Zhao, S.; Wang,
Y.; Kopinja, J.; Qu, X.; Reilly, J.; Walker, D.; Xu, L.; Aleksandrowicz, D.; Marshall,
C. G.; Scott, M. L.; Kohl, N. E.; Bachman, E. Biochem. Pharmacol. 2009, 78, 382.
9. Compounds 1, 14, and 31 were tested against a broad panel of 98 kinases at
Invitrogen. Selectivity folds were calculated based on estimated IC50s from 3
point titrations (1, 100, 1000 nM) and the internal JAK2 potencies.
10. Cheng, A. C.; Eksterowicz, J.; Geuns-Meyer, S.; Sun, Y. J. Med. Chem. 2010, 53,
4502.
Figure 5. Inhibitor 24 modeled in ATP binding site of JAK2 enzyme highlighting
potential interaction with Asn-981.
modeling studies provided valuable insights into the improved ki-
nase selectivity. These results, together with monitoring physico-
chemical properties and lipophilicity, led to the discovery of 31, a
kinome selective JAK2 inhibitor with favorable in vivo properties.
11. For full experimental procedure see WO2009035575.
12. Compounds were docked into a crystal structure of JAK2 (Chain A of PDB ID
2B7A17) using GLIDE v58515 (Schrödinger, LLC) in XP mode using default
parameters. During docking, the compound was required to make hydrogen
bonds to the backbone of residues Glu-930 and Leu-932. Resulting poses were
subsequently filtered through visual inspection. The protein structure was
prepared for calculation using the PrepWizard functionality within MAESTRO
v9.3.5 (Schrödinger, LLC) retaining chain A and water molecule 139 as the
receptor.
Acknowledgments
T.S. thanks Kerrie Spencer, Meredeth McGowan, Jongwon Lim,
and Andrew Haidle for careful reading of this manuscript and
Bruce Adams for NMR support.
13. Lucet, I. S.; Fantino, E.; Styles, M.; Bamert, R.; Patel, O.; Broughton, S. E.; Walter,
M.; Burns, C. J.; Treutlein, H.; Wilks, A. F.; Rossjohn, J. Blood 2006, 107, 176.
14. (a) Finlay, M. R. V.; Acton, D. G.; Andrews, D. M.; Barker, A. J.; Dennis, M.;
Fisher, E.; Graham, M. A.; Green, C. P.; Heaton, D. W.; Karoutchi, G.; Loddick, S.
A.; Morgentin, R.; Roberts, A.; Tucker, J. A.; Weir, H. M. Bioorg. Med. Chem. Lett.
2008, 18, 4442; (b) Barillari, C.; Duncan, A. L.; Westwood, I. M.; Blagg, J.; van
Montfort, R. L. M. Proteins: Struct., Funct., Bioinf. 2011, 79, 2109.
15. Van de Waterbeemed, H.; Smith, D. A.; Jones, B. C. J. Comput. Aided Mol. Des.
2001, 15, 273.
References and notes
1. Vainchenker, W.; Constantinescu, S. N. Oncogene 2013, 32, 2601.
2. Reddy, M. M.; Deshpande, A.; Sattler, M. Expert Opin. Ther. Targets 2012, 16, 313.
3. (a) Antonysamy, S.; Hirst, G.; Park, F.; Sprengler, P.; Stappenbeck, F.; Steensma,
R.; Wilson, M.; Wong, M. Bioorg. Med. Chem. Lett. 2009, 19, 279; (b) Burns, C. J.;
Bourke, D. G.; Andrau, L.; Bu, X.; Charman, S. A.; Donohue, A. C.; Fantino, E.;
Farrugia, M.; Feutrill, J. T.; Joffe, M.; Kling, M. R.; Kurek, M.; Nero, T. L.; Nguyen,
T.; Palmer, J. T.; Phillips, I.; Shackleford, D. M.; Sikanyika, H.; Styles, M.; Su, S.;
Treutlein, H.; Zeng, J.; Wilks, A. F. Bioorg. Med. Chem. Lett. 2009, 19, 5887; (c)
Ioannidis, S.; Lamb, M. L.; Wang, T.; Almeida, L.; Block, M. H.; Davies, A. M.;
Peng, B.; Su, M.; Zhang, H.-J.; Hoffmann, E.; Rivard, C.; Green, I.; Howard, T.;
Pollard, H.; Read, J.; Alimzhanov, M.; Bebernitz, G.; Bell, K.; Ye, M.; Huszar, D.;
Zinda, M. J. Med. Chem. 2011, 54, 262; (d) Lim, J.; Taoka, B.; Otte, R. D.; Spencer,
16. Compounds 24 and 31 were profiled as racemates.
17. Katz, J. D.; Lapointe, B. T.; Dinsmore, C. J. Org. Chem. 2009, 22, 8866.
18. Compounds were dosed by oral gavage to C56bl/6 mice as a solution in
40%PEG400: 50%Trappsol: 10% water. Drug concentrations and pSTAT5 signal
were measured at 1, 3, and 8 h.